Collects bounded packet captures from AKS nodes and Azure network configuration for wire-level evidence. WHEN: "capture packets on an AKS node", "take a pcap", "run tcpdump on AKS", "prove where packets drop". Use for explicit packet-capture intent after read-only diagnostics, not general AKS connectivity or ingress troubleshooting.

Set up AI Runway on AKS — from bare cluster to running model. Covers cluster verification, controller install, GPU assessment, provider setup, and first deployment. WHEN: "setup AI Runway", "onboard AKS cluster", "install AI Runway", "airunway setup", "deploy model to AKS", "GPU inference on AKS", "KAITO setup on AKS", "run LLM on AKS", "vLLM on AKS", "set up model serving on AKS", "AI Runway controller".

Create commits, pull requests, and GitHub PR reviews. Use when the user wants to commit changes, open or refresh a pull request, or add/submit PR review comments.

Runs a product quality review on a journey, feature, landing page, onboarding flow, or AI/agent surface and returns a friction log, a scored critical-journey review, and a taste bar your team can calibrate against. Use when someone asks is this good enough to ship, how polished should it be, why do users bounce, what is our quality bar, how do we build taste, or to review a design, prototype, onboarding, or checkout. Draws on 264 insights from 100 Lenny's Podcast guests incl. Katie Dill, Dylan Field and Stewart Butterfield.

Use the diffs tool to produce real, shareable diffs (viewer URL, file artifact, or both) instead of manual edit summaries.

Curate noteworthy Docker launches from documentation pull requests merged during a requested period. Use when generating or reviewing data/whats-new.json, preparing the Docker Docs What's new timeline, or deciding which documented releases merit Docker-wide highlights.

Decide whether and how errors report to Sentry. Use when touching capture or error-handling paths in `web/**`, triaging Sentry noise, or changing Sentry settings.

Execute quality gate validation with configurable blocking behavior. Use when running type-check, build, tests, lint, or custom validation commands in orchestrators or workers to enforce quality standards.

Use this skill whenever delegating work to subagents — whether a single Task call or a full team via TeamCreate. Guides delegation, task decomposition, agent selection, and coordination. Covers when to spawn architect (design decisions, trade-offs), researcher (explorations, technical investigation), developer (FEATURE specs and code), tech-writer (documentation), qa (post-implementation validation), and architecture-critic (stress-testing architect output). Trigger on any subagent spawn, team creation, multi-agent coordination, or parallel work request.

Research competitor products, launch history, and acquisition evidence from public sources or supplied materials. Use when evaluating growth claims, comparing competitors, or choosing growth experiments. Produces a cited brief that separates observations, attributed claims, estimates, and hypotheses.

Revalidate persistent agent knowledge by classifying why it can become stale and how quickly it changes, then retain, revise, externalize, or retire it. Use when reviewing skills, stored instructions, facts, preferences, assumptions, or workflows; when an owner, decision, or dependency changes; or when self-improvement captures, reviews, or promotes a learning. Works independently of self-improvement. Not for runtime context-window degradation, unsaved task observations, or age-based deletion.

Adds a new LLM provider implementing LLMProvider interface with call() and stream() methods. Integrates with provider factory in src/llm/index.ts, config detection in src/llm/config.ts, and error handling via tracking and recovery. Use when adding a new model backend, integrating a third-party LLM API, or extending LLM platform support. Do NOT use for fixing bugs in existing providers, modifying existing provider behavior, or changing the LLMProvider interface.

Adds a new LLM provider implementing LLMProvider interface with call() and stream() methods. Integrates with provider factory in src/llm/index.ts, config detection in src/llm/config.ts, and error handling via tracking and recovery. Use when adding a new model backend, integrating a third-party LLM API, or extending LLM platform support. Do NOT use for fixing bugs in existing providers, modifying existing provider behavior, or changing the LLMProvider interface.

Generate django-modern-rest transport-layer skeletons from OpenAPI 3.1+ specs. Use when asked to turn an OpenAPI file, URL, or pasted document into typed DTOs, controllers, routers, Django URL wiring, and minimal tests for this repository or similar projects built on dmr. Trigger on requests to scaffold APIs, bootstrap apps, or map OpenAPI operations to Controller, Router, and OpenAPI view constructs without implementing business logic.

Prepare a sourced Minutes meeting outcome for human-governed closeout of existing X1 work. Use when the user wants a meeting decision or commitment to close, defer, escalate, or retire a specific X1 coordination thread. Never use it to infer settlement, move money, contact anyone, or close X1 work automatically.

Implements a feature or bug fix test-first: write one failing test, run it and watch it fail, write the minimal code to pass, then refactor with the tests green. Use when a change alters code behavior: a new feature, a bug fix, or a behavior change, whether the request comes alone or through delivery-flow.

Post-translational modification (PTM) analysis — phosphorylation, ubiquitination, acetylation, glycosylation, methylation. Uses iPTMnet (sites + enzymes), ProtVar (functional consequences), UniProt (baseline), STRING, ELM (linear motifs), MassIVE/ProteomeXchange (experimental), and the EBI Proteins API (mutagenesis, MS-proteomics/HPP peptide evidence, antigenic regions, genomic coordinates, protein-level variant lookup by accession/dbSNP/HGVS) as a cross-validation source. Use for PTM site annotation, kinase-substrate identification, and PTM-disease associations.

Turns the current session into the operator's SDLC proxy for parallel PRP workstreams in isolated worktrees. It owns the combined outcome, steers autonomous deliveries, holds human and merge gates, verifies proof, and sequences merges. Use when the user wants to "spawn N agents in separate worktrees", "run prp-issue on these issues in parallel", "orchestrate these features", "act as my orchestrator", "coordinate agents through the PRP pipeline", "ship these issues in parallel", or invokes /prp-orchestrate.

Turns the current session into the operator's SDLC proxy for parallel PRP workstreams in isolated worktrees. It owns the combined outcome, steers autonomous deliveries, holds human and merge gates, verifies proof, and sequences merges. Use when the user wants to "spawn N agents in separate worktrees", "run prp-issue on these issues in parallel", "orchestrate these features", "act as my orchestrator", "coordinate agents through the PRP pipeline", "ship these issues in parallel", or invokes /prp-orchestrate.

Make any UI appear as if an AI agent is generating it. A dotted wireframe is drawn from the element's own structure (outline, boxes, text lines, icon paths, with guides, corner handles and size labels), then the real UI is revealed behind a scan line with a fade on a smooth ease-out. Two modes, skeleton and skeleton + generation texts (an agent log, the motion CSS being written, and a step line). One dependency-free runtime that works with any framework (vanilla, React, Next.js, Vue, Svelte). Use when the user asks for a "generation animation", "AI generating the UI", "agent building the component", "build-in animation", "wireframe to UI reveal", "skeleton generation", "dotted wireframe", "matrix dots", "make it look generated", "futuristic reveal", "generative UI loading", or for an AI product that should show its output being built. Also the commands transitions generate, transitions generate texts and transitions generate preview.